Recent research has indicated that significant numbers of opiate dependent patients have serious, coexisting psychiatric disorders and that the addition of time-limited psychotherapy to treatment- as-usual leads to significant improvements which are evident even 12 months following treatment (Woody et al., 1983). It is now important to understand the effective ingredients of psychotherapy for opiate addiction as a means towards refining and improving treatment so as to maximize the benefits addicts received. It is the aim of the proposed research to apply new psychotherapy process methods to the sessions of a recent project by our group evaluating the usefulness of Supportive-Expressive Psychodynamic psychotherapy and Cognitive-Behavioral psychotherapy in the treatment of opiate addiction. The application of measures from the psychodynamic perspective and the cognitive perspective will allow us to test and compare two major models of the effective ingredients of change. In dynamic therapy, it is proposed that the accuracy of the therapist's interpretations is the central variable, and in cognitive therapy it is proposed that the amount and quality of cognitive therapy provided leads to changes in causal explanations which leads to better outcome. Measures of these concepts will be applied to three early-in-treatment sessions and three late-in-treatment sessions from the treatments of 59 opiate addicts who participated in the Woody et al. (1983) outcome study. We will examine the predictive value of each variable in the context of variables from the other model and also controlling for established predictor variables (e.g., the strength of the therapeutic alliance). Session-by-session changes will be analyzed in a panel regression approach to tease out spurious explanations of correlational findings.